Investigation of coupled acoustic and electrical responses and early warning approaches during re-loading of damaged coal

• Initial damage from engineering disturbances significantly reduces the load-bearing capacity of coal, as evidenced by decreased resistivity and increased acoustic emission counts during reloading. • Time-frequency analysis reveals that acoustic emission spectra evolve from bimodal to broadband with enhanced low- and high-frequency components, while resistivity spectra show bandwidth broadening and high-frequency advancement. • 3D AE localization and fractal-dimension analysis indicate that the spatial fractal dimension of cracks rises significantly during compaction, providing insights into damage evolution. • An integrated early-warning model combining multiscale entropy fusion, TCN-Transformer forecasting, recurrence-network analysis, and Bayesian framework achieves a 28.4-second lead time for dynamic hazard prediction.